On-line soft error correction in matrix–matrix multiplication

Soft errors are one-time events that corrupt the state of a computing system but not its overall functionality. Soft errors normally do not interrupt the execution of the affected program, but the affected computation results cannot be trusted any more. A well known technique to correct soft errors in matrix–matrix multiplication is algorithm-based fault tolerance (ABFT). While ABFT achieves much better efficiency than triple modular redundancy (TMR) – a traditional general technique to correct soft errors, both ABFT and TMR detect errors off-line after the computation is finished. This paper extends the traditional ABFT technique from off-line to on-line so that soft errors in matrix–matrix multiplication can be detected in the middle of the computation during the program execution and higher efficiency can be achieved by correcting the corrupted computations in a timely manner. Experimental results demonstrate that the proposed technique can correct one error every ten seconds with negligible (i.e. less than 1%) performance penalty over the ATLAS dgemm().     

Rapid determination of yunaconitine and related alkaloids in aconites and aconite‐containing drugs by ultra high‐performance liquid chromatography–tandem mass spectrometry

Yunaconitine (YAC) is a toxic aconite alkaloid that is considered to be a hidden aconite poison since it is frequently found in body fluids from aconite poisoning patients, but has not been well studied in commonly used herbal drugs. In this paper, a rapid and sensitive ultra high‐performance liquid chromatography–tandem mass spectrometry (UHPLC‐MS/MS) detection combined with microwave‐assisted extraction (MAE) was developed for high throughput simultaneous determination of YAC and six other toxic aconite alkaloids in 31 samples of crude, processed aconites and aconite‐containing drugs. The optimized method showed excellent linearity, precision, accuracy and recovery for all target compounds with short run time. YAC was detected in some samples with contents from 0.015 to 10.41 mg/g. This is the first report on the determination of YAC in Radix Aconiti, Radix Aconiti Kusnezoffii and aconite‐containing drugs. This newly developed method facilitates the rapid screening of YAC and related toxic aconite alkaloids and allows YAC to be used as a chemical marker for the quality control of aconites and aconite‐containing drugs. Copyright © 2012 John Wiley & Sons, Ltd.     

Elucidating ‘undecipherable’ chemical structures using computer‐assisted structure elucidation approaches

Structure elucidation using 2D NMR data and application of traditional methods of structure elucidation are known to fail for certain problems. In this work, it is shown that computer‐assisted structure elucidation methods are capable of solving such problems. We conclude that it is now impossible to evaluate the capabilities of novel NMR experimental techniques in isolation from expert systems developed for processing fuzzy, incomplete and contradictory information obtained from 2D NMR spectra. Copyright © 2012 John Wiley & Sons, Ltd.     

Optimized microwave‐assistant extraction combined ultrasonic pretreatment of flavonoids from Periploca forrestii Schltr. and evaluation of its anti‐allergic activity

Microwave extraction combined ultrasonic pretreatment of flavonoids from Periploca forrestii Schltr. was investigated in this study, extraction process was first performed in an ultrasonic cleaner, then treated by microwave irradiation. The optimum ultrasonic time of 25 min was selected by single‐factor experiments. A response surface methodology has been used to obtain a mathematical model that describes the process and analyzes the significant parameters ethanol concentration 59.92%, liquid to raw materials ratio 21.24 mL/g, microwave radiation time 209.53 s, and microwave power 274.14 w. In these optimum conditions, the yield of flavonoids from P. forrestii (TFPF) could be up to 9.11 ± 0.08%, which was increased by 14.30 and 19.86% compared microwave extraction and ultrasonic extraction, respectively. In vitro suppress hyaluronidase experimentation showed that TFPF purified using polyamide exhibited good anti‐hyaluronidase ability with IC₅₀ value of 1.033 mg/mL, possessing certain anti‐antiallergic and potential application prospect in pharmaceutical production of treating inflammation and other related fields.     

Zinc oxide nanorod‐loaded activated carbon for ultrasound‐assisted adsorption of safranin O: Central composite design and genetic algorithm optimization

This paper focuses on the development of an effective methodology to obtain the optimum ultrasonic‐assisted removal of a dye, safranin O (SO), under optimum conditions that maximize the removal percentage, using ZnO nanorod‐loaded activated carbon (ZnO‐NRs‐AC) in aqueous solution. Central composite design coupled with genetic algorithm was used for parameter optimization. The effects of variables such as pH, initial dye concentration, mass of ZnO‐NRs‐AC and sonication time were studied. The interactive and main effects of these variables were evaluated using analysis of variance. The structural and physicochemical properties of the ZnO‐NRs‐AC adsorbent were investigated using field emission scanning electron microscopy and X‐ray diffraction. Adsorption equilibrium data were fitted well with the Langmuir isotherm and the maximum monolayer capacity was found to be 32.06 mg g^?1. Studies of the adsorption kinetics of the SO dye showed a rapid sorption dynamic with a ^pseudo‐second‐order kinetic model, suggesting a chemisorption mechanism.     

Continuous Generalized Gradient Descent

This article derives characterizations and computational algorithms for continuous general gradient descent trajectories in high-dimensional parameter spaces for statistical model selection, prediction, and classification. Examples include proportional gradient shrinkage as an extension of LASSO and LARS, threshold gradient descent with right-continuous variable selectors, threshold ridge regression, and many more with proper combinations of variable selectors and functional forms of a kernel. In all these problems, general gradient descent trajectories are continuous piecewise analytic vector-valued curves as solutions to matrix differential equations. We show the monotonicity and convergence of the proposed algorithms in the loss or negative likelihood functions. We prove that approximations of continuous solutions via infinite series expansions are computationally more efficient and accurate compared with discretization methods. We demonstrate the applicability of our algorithms through numerical experiments with real and simulated datasets.     

A review on enzyme and ultrasound: A controversial but fruitful relationship

A critical review on the effect of ultrasound (US) on enzymes and their biocatalytic action is presented here. Discussion on the information users of US acquire before utilizing the different devices, and the importance they give to US frequency is constant along the review. The authors have gone into the different areas in which the US–enzyme binomial has been applied. The lack of enough information on the US–enzyme-working conditions under which each piece of research has been developed, and the necessity to provide complete information on the data and metadata to give enough light on each piece of research (and thus on the potential comparison of results from different studies) are critically exposed. With this aim, the study has been divided into the positive effect of US on enzymes to favor the production of metabolites, polymers or proteins; and the degradation, inhibition or activation of the biocatalyst under US application. Also the effect of US on enzyme production and the main fields of application of the US–enzyme binomial are discussed.     

High‐throughput salting‐out assisted liquid–liquid extraction with acetonitrile for the determination of anandamide in plasma of hemodialysis patients with liquid chromatography tandem mass spectrometry

Anandamide (AEA) is an endocannabinoid present in human plasma that is associated with several physiological functions and disease states. However, low AEA plasma levels pose challenges in terms of analytical characterization. Classical liquid‐based lipid extraction and solid‐phase extraction require complicated procedures and the drying down of relatively large volumes of solvents, making them unsuitable for high‐throughput analysis. Here a high‐throughput salting‐out assisted liquid–liquid extraction (SALLE) method with acetonitrile and mass spectrometry compatible salts for liquid chromatography–tandem mass spectrometry (LC‐MS/MS) analysis of AEA in human plasma has been developed and validated. The seamless interface of SALLE and LC‐MS eliminated the drying‐down step, only 100 μL of plasma is required and minimal volumes of organic solvent are used. Good reproducibility, accuracy and precision were demonstrated during the method validation. The method is linear up to 10 ng/mL with a lower limit of quantitation of 0.1 ng/mL for AEA, the accuracy for AEA was from 93.3 to 96.7% and the precision was <8.57%. This new methodology was successfully applied to analysis of clinical samples from maintenance hemodialysis patients. Copyright © 2015 John Wiley & Sons, Ltd.     

Color restoration in the black-and-white video camera

A method of restoring the color information in a black-and-white video camera is proposed in this paper. The method is demonstrated by an experiment. The experiment contains two steps. Firstly, an environmental coefficient matrix is calculated by taking pictures on a calibration plate. Then, a restored picture is estimated by Matlab. The experiments have been done in different settings and in another dual-mode camera. All restored pictures are compared with the original ones. The results indicate that this method has precision, repeatability and adaptability to environment. If the system can be put to practical use, it will give clues to the criminal investigation.     

Ultrasound assisted co-precipitation of nanostructured CuO–ZnO–Al2O3 over HZSM-5: Effect of precursor and irradiation power on nanocatalyst properties and catalytic performance for direct syngas to DME

Nanostructured CuO–ZnO–Al₂O₃/HZSM-5 was synthesized from nitrate and acetate precursors using ultrasound assisted co-precipitation method under different irradiation powers. The CuO–ZnO–Al₂O₃/HZSM-5 nanocatalysts were characterized using XRD, FESEM, BET, FTIR and EDX Dot-mapping analyses. The results indicated precursor type and irradiation power have significant influences on phase structure, morphology, surface area and functional groups. It was observed that the acetate formulated CuO–ZnO–Al₂O₃/HZSM-5 nanocatalyst have smaller CuO crystals with better dispersion and stronger interaction between components in comparison to nitrate based nanocatalysts. Ultrasound assisted co-precipitation synthesis method resulted in nanocatalyst with more uniform morphology compared to conventional method and increasing irradiation power yields smaller particles with better dispersion and higher surface area. Additionally the crystallinity of CuO is lower at high irradiation powers leading to stronger interaction between metal oxides. The nanocatalysts performance were tested at 200–300 °C, 10–40 bar and space velocity of 18,000–36,000 cm³/g h with the inlet gas composition of H₂/CO = 2/1 in a stainless steel autoclave reactor. The acetate based nanocatalysts irradiated with higher levels of power exhibited better reactivity in terms of CO conversion and DME yield. While there is an optimal temperature for CO conversion and DME yield in direct synthesis of DME, CO conversion and DME yield both increase with the pressure increase. Furthermore ultrasound assisted co-precipitation method yields more stable CuO–ZnO–Al₂O₃/HZSM-5 nanocatalyst while conventional precipitated nanocatalyst lost their activity ca. 18% and 58% in terms of CO conversion and DME yield respectively in 24 h time on stream test.